> I}@bjbjWW>H==x_8RR!!;";";"O"O"O"8"s#TO"\;h%w(:(((*.**:::::::$=v@~):;"*N*@***):!!((;444*&!8(;"(:4*:44V/9@#"9(V0o99,;0\;{9T@1@9@;"9 **4*****):):4***\;****@*********R? :S. Bryzgalova,S. Gelman
State University Higher School of Economics
SPURIOUS LONG-RANGE DEPENDENCE IN EMERGING FINANCIAL MARKETS
1. Introduction
Long memory in financial markets has been a widely discussed issue for the last 40 years. Originally found by Mandelbrot and Wallis (1968), significant autocorrelation in the yields of various financial instruments at low frequency, is one of its manifestations. Further research confirmed that this is not the only paradoxal evidence from the memory of the process the autocorrelation function itself decays much slower than the one corresponding to standard ARMA GARCH processes. Long memory presence has been recently extensively used to test the market efficiency and stage of development, for it allows for significant predictability [Rogers, 1997; Eom et al., 2008] and therefore, serves as an indicator of market inefficiency, often applied to emerging markets [Dacorogna et al., 2005; Cajueiro, Tabak, 2006; Eom et al., 2008].
The problem with estimating the long range dependence is that a process with structural breaks generates exactly the same statistical properties as the long memory model, and there is still no test on their differentiation (see Section 3). However, this problem is often neglected in the empirical studies.
We suggest using predictive power of the model to give evidence in favor of the true underlying process. The logic is straightforward: since the long range dependence is supposed to yield superior predictive abilities, we should test them against those of a regime-switching process. For this purpose we consider emerging markets of BRIC countries. In particular, we estimate ARFIMA and Markov Switching model, build a series of one-step ahead forecasts and them compare them with Granger Newbold and Diabold Mariano tests. We show that MS process demonstrates superior predictive powers compared to those of ARFIMA for all the markets. Hence, it is quite likely that traditional long memory estimations are spurious, being due to structural breaks.
The paper is organized as follows. Section 2 discusses the properties of long memory processes and their application in finance, section 3 focuses on the dichotomy between long-range dependence and structural breaks. Section 4 describes the used methodology, section 5 provides the data description. Finally, section 6 presents the results of the estimation, and section 7 concludes, summarizing findings and outlining directions for future research.
2. Motivation for long memory
Significant autocorrelation at low frequencies and a slow decay of the autocorrelation functions are two basic manifestations of the long memory process. Theoretically these unusual patterns could be explained by the heterogeneity of the agents present on the market. In the recent decade there have been proposed a number of models, directly incorporating it into the market structure [Brock, Hommes, 1998; LeBaron, 2000; Abel, 2008]. Finally, there have been developed Heterogeneous Market Hypothesis [Dacorogna et al., 2001] which provides a general framework for these models and postulates major characteristics of the corresponding market structure.
According to HMH, the market is populated by heterogeneous agents of different time horizons and dealing frequency, each having the short-term memory with a standard exponential decay (in different time scales). It is a known mathematical fact that the superposition of different exponential decays at various scales results in the behaviour close to a hyperbole. Hence, such a market structure could perfectly explain the hyperbolic decay of autocorrelation widely observed in real financial markets.
Long memory processes allow for significant predictability [Rogers, 1997; Eom et al., 2008] and therefore, its presence can be used as an indicator of market inefficiency. Introducing the long memory components directly into the asset pricing theory is rather difficult, but there has emerged a considerable area of research on derivative pricing, namely options, which accounts for possible long memory in stock returns [Jamdee, Los, 2007]. It has been demonstrated that the modified approach results in much better explanation of the market realizations than the standard Black and Scholes formulation.
3. Long memory VS structural breaks
Structural breaks have long been an integral feature of many financial time series. The pioneering work belongs to Chow (1960), who suggested an F-test for a linear regression model, provided the known date of the break. The limitation of exogeneity was overcome in further studies for endogenous structural breaks: Brown, Durbin and Evans (1975) proposed a procedure based on the cumulative sum of the residuals, known as CUSUM-test. Later studies accounted for possible heteroscedasticity and multiple structural breaks [Andrews, 1993; Andrews, Ploberger, 1994; Bai, Perron, 1998].
Several studies were conducted to check tests validity in the presence of the long memory. In particular, Kramer and Sibbertsen (2000) proved that in the presence of long memory, the null distribution of CUSUM statistics tends to infinity, thus, detecting a spurious structural break. Alas, not only a long memory process can successfully imitate the properties of structural breaks, but the vice versa is also possible. In particular, there have been a number of research papers on the latter effect, including [Ding, Granger, 1996; Andersen, Bollerslev, 1997; Granger, Hyung, 2004; Diebold, Inoue, 2001; Choi, Zivot, 2007].
A new research area emerged with the introduction of Markov Switching processes by Hamilton (1988). At each point of time the process can be in one of the specified states of nature, and not only do we estimate each regime parameters, but also the probabilities to be in corresponding states at each moment of time. This idea was used in other specifications of the regime switching models, such as Diebold and Inoue (2001), or Granger and Hyung (2004). These processes can perfectly explain the autocorrelation patterns usually prescribed to long memory evidence.
It is still an open question, how to differentiate between the long memory and regime switching processes. While we do not suggest a formal procedure, we think that in case of financial time series predictive power of the models can be regarded as a valuable criterion. In particular, the main implication of long memory estimations consists in predicting stock returns and testing Efficient Market Hypothesis. Hence, a true long memory process should exhibit a substantial predictive power, which we can test against that of the regime switching process.
4. Methodology
4.1. ARFIMA-model
With a fractional integration parameter d, the ARFIMA(p,d,q) model is written as
EMBED Equation.3
where L stands for the lag operator, (L) and (L) are polynomials of order p and q correspondingly.
The simplest specification of this model if the so-called fractional Brownian motion (FBM):
QUOTE EMBED Equation.3 with EMBED Equation.3 QUOTE
It is easy to note through a simple factorization why this process exhibits long memory:
EMBED Equation.3
where stands for gamma-function.
The autocorrelation function for this process demonstrates hyperbolic decay and looks as follows:
EMBED Equation.3
An algorithm for the computation of the autocovariance function of a general ARFIMA process is derived in [Sowell, 1992].
We consider three methods to estimate the parameters of ARFIMA:
Exact Maximum Likelihood, as in [Ibid],
Modified Profile Likelihood, as in [An, Bloomfield, 1993], and
Nonlinear Least Squares.
4.2. Markov switching (MS) model
The MS model has been developed by Hamilton (1988). Consider the following AR(1) model
EMBED Equation.3
At every period of time the process belongs to one of the regimes, which are characterized by different coefficients and possibly different variance of the disturbances. The state variable y(t) is assumed to follow an ergodic first-order Markov process and is characterized by the matrix of transition probabilities QUOTE EMBED Equation.3 from state i to state j.
EMBED Equation.3
Once the coefficients of the model and the transition matrix have been estimated, the probability of being in state j, based on the knowledge of complete series
EMBED Equation.3 QUOTE can be calculated for each date (the so-called smoothed pro-
babilities), as described in [Kim, Nelson, 1999]. When only information up to a cur-
rent date is used, the probability estimation EMBED Equation.3 QUOTE is called the filter probability.
To estimate the model we use (quasi) maximum likelihood method [Hamilton, 1988].
4.3. Tests for predictive power
In order to test the predictive powers of one model against the other we use Granger Newbold (1976) and Diebold Mariano (1995) tests for 1-step ahead predictions.
Consider 2 models estimated at the same data sample, generating the forecasts EMBED Equation.3 QUOTE Forecast errors are defined as
EMBED Equation.3
Granger Newbold test.
Let us form new sequences EMBED Equation.3 QUOTE and EMBED Equation.3 QUOTE
Provided that the forecast errors are unbiased and normally distributed, if both models produce equally good forecasts, then x(t) and z(t) should be uncorrelated, and the statistics looks as follows:
EMBED Equation.3
where EMBED Equation.3 QUOTE is a sample correlation of x(t) and z(t), H is the number of forecasts generated by each model.
Diabold Mariano test.
This procedure uses a more general setting to test the predictive power of the models, for it does not require the forecast errors to be unbiased, normally distributed or serially uncorrelated.
The mean loss is defined as
EMBED Equation.3
If the loss time series EMBED Equation.3 QUOTE is autocorrelated of order q, then provided that both forecasts are equally adequate,
EMBED Equation.3
As far as the functional form of g() is concerned, a common approach is to use the square.
5. Data
We conduct our analysis using the data from national indexes of BRIC countries.
BOVESPA of Brazil. The Bovespa Index is the main indicator of the Brazilian stock markets average performance. The issuing companies of the stocks that compose the Bovespa Index theoretical portfolio are responsible, in average, for approximately 70% of the sum of all BOVESPAs companies capitalization.
MICEX of Russia. The MICEX Index is a capitalization-weighted index of the market of most liquid 19 stocks of Russian issuers admitted to circulation on the MICEX (Moscow Interbank Currency Exchange).
India BSE (100) National for India. India BSE(100) is calculated using the Free-float Market Capitalization methodology. The equity shares of 100 companies from the Specified and the Non-Specified list of the five major stock exchanges, viz. Mumbai, Calcutta, Delhi, Ahmedabad and Madras have been selected for the purpose of compiling the BSE National Index.
Shanghai SE Composite for China. Constituents for SSE Composite Index are all listed stocks (A shares and B shares) at Shanghai Stock Exchange.
We have chosen the time interval of 01.200001.2008, since it was rather stable for the countries of interest (no periods of hyperinflation or financial crisis). Log-returns are then computed and used for the analysis.
6. Estimation
An essential feature of ARFIMA-model consists in the ability to capture both short- and long-range effects of investors reaction towards news. In addition, autoregressive function for ARFIMA-model demonstrates not exponential, but hyperbolic rate of decay, which can account for significant autocorrelation at high lags.
We chose the model on the basis of information criteria and autocorrelation fit. The estimation was done in OxMetrics 4,0, results are summarized in Table 1.
Table SEQ Table \* ARABIC 1.ARFIMA(p,d,q) estimates
CountryCoefficientsExact Maximum LikelihoodModified Profile LikelihoodNonlinear Least Squares Coefficient St.error Coefficient St.error Coefficient St.errord** 0,042528 0,01776d** 0,06291 0,01785d* 0,0394184 0,02881AR-5* 0,042037 0,02206AR-5* 0,041956 0,02207AR-5* 0,0411425 0,02248AR-6 0,003650 0,02195AR-6 0,003775 0,02196AR-6 0,0063301 0,02219AR-7* 0,036474 0,02194AR-7* 0,036313 0,02196AR-7* 0,0296034 0,02211AR-8 0,019318 0,02194AR-8 0,019134 0,02196AR-8 0,014776 0,02201AR-9* 0,031507 0,02191AR-9* 0,031309 0,02193AR-9* 0,0262247 0,02194AR-10** 0,059842 0,02196AR-10** 0,06006 0,02197AR-10** 0,0620966 ,02201BrazilMA-1* 0,0393780 0,0363020002008AIC 5,20955043AIC 5,20955043AIC 5,20955043 Coefficient St.error Coefficient St.error Coefficient St.errord** 0,164521 0,08484d* 0,137394 0,07922d** 0,167749 0,08379AR-1** 0,178968 0,08876AR-1* 0,152034 0,08294AR-1* 0,184778 0,08780AR-2* 0,054598 0,03608AR-2* 0,0456001 0,03594AR-2* 0,0519264 0,03550AR-3 0,0079348 0,02835AR-3 0,0134459 0,02841AR-3 0,0055267 0,02823AR-4** 0,0629145 0,02894AR-4** 0,0571543 0,02865AR-4** 0,0601306 0,02866RussiaAR-9** 0,0515149 0,02401AR-9** 0,0486693 0,02374AR-9** 0,0520668 0,0238520002008AIC 4,73139209AIC 4,71767168AIC 4,752945 Coefficient St.error Coefficient St.error Coefficient St.errord* 0,0232635 0,02193d 0,0165652 0,03583d 0,0077324 0,04165AR-1** 0,119743 0,04388AR-1** 0,113240 0,04354AR-1 0,0008142 0,04352AR-4** 0,062277 0,02350AR-4** 0,0612208 ,02346AR-4** 0,0562914 0,02340AR--9** 0,0465497 0,02225AR-9** 0,0463685 0,02226AR-9** 0,0447982 0,02211IndiaAR-10* 0,0360177 0,02198AR-10* 0,0362668 0,02200AR-10* 0,0422150 0,0243820002008AIC 5,3351933AIC 5,3201407AIC 5,34626565Continued
CountryCoefficientsExact Maximum LikelihoodModified Profile LikelihoodNonlinear Least Squares Coefficient St.error Coefficient St.error Coefficient St.errord* 0,118024 0,06802d* 0,118046 0,06805d* 0,112709 0,06684AR-1*** 0,583834 0,1365AR-1*** 0,583824 0,1365AR-1*** 0,556809 0,1399AR-4** 0,057311 0,02594AR-4** 0,057308 0,02594AR-4** 0,060619 0,02579AR-5*** 0,083117 0,02278AR-5*** 0,083118 0,02279AR-5*** 0,080423 0,02232AR-10* 0,031185 0,02237AR-10* 0,031183 0,02238AR-10* 0,0290795 0,02198ChinaMA-1 0,0690887 0,1074MA-1 0,06909 0,1075MA-1 0,0665491 0,113320002008AIC 5,60483872AIC 5,59682986AIC 5,62015626
Emerging markets are characterized by long memory (significant value of d at least at 10%), which is regarded by the previous researchers as a sign of market inefficiency. Using similar estimates, they described the current stage of development in countries like BRIC as relatively low and the degree of predictability as rather substantial.
Table 2 summarizes MS-estimation results for the same data.
Table 2.Markov switching estimates
CountryModel SpecificationTransition matrixBrazil EMBED Equation.DSMT4 EMBED Equation.DSMT4 Russia EMBED Equation.DSMT4 EMBED Equation.DSMT4 India EMBED Equation.DSMT4 EMBED Equation.DSMT4 China EMBED Equation.DSMT4 EMBED Equation.DSMT4 Although not all the coefficients in the estimated model are significant, the overall tendency is that the algorithm differentiates between positive and negative dynamics on the market, the latter characterized by a substantial increase in volatility.
Having estimated the models, we divide our sample in 2 parts, taking 7 years for the estimation period and 1 year (250 observations) for the forecast period. Note, that it is rather short compared to the dominating approach in the literature (about 1/3 of the sample). However, robust long memory estimation requires a very large data set; hence, we had to allow only for 250 observations. Each period we estimate models and make 1-step ahead forecast, then update the information and repeat the procedure. This yields 250 static forecasts for both models.
Finally, we carry out the tests on the predictive power of our models.
Table 3.Granger Newbold test
BrazilRussiaIndiaChinaGN-statistics3,244510,2886731,50061,6777HypothesisH1(1%)H0H0H1(10%)
For Brazil and China we tend to reject the null hypothesis of equal predictive power. However, for Russia and India the statistics lies within the confidence interval. However, Granger Newbold test is rather restrictive, for it assumes unbiased normally distributed forecast errors with no serial autocorrelation, while Diabold Mariano test is free from these drawbacks. To conduct this test, we assumed that the first model is ARFIMA, and the second MS, so that a strictly positive value of the statistics means worse predictive power of the former.
Table 4.Diebold Mariano test for different orders of autocorrelation included
qBrazilRussiaIndiaChinaq = 18,8045637,7347375,858465226,911763q = 27,3894387,2581424,912317737,109308q = 3 6,7402475,7067794,451016736,427741q = 46,498625,4258334,074562394,057996q = 56,207975,1497084,017625695,482122q = 66,0557834,8847364,025614985,66679Continued
qBrazilRussiaIndiaChinaq = 75,902064,8141114,00166265,845283q = 85,0297224,60454,0050255,936899q = 95,7980054,4707873,999793975,719408q = 105,9698354,4166243,985229415,491889
Regardless of the number of autocorrelation coefficients included MS models demonstrate superior predictive power to that of the ARFIMA-model for any country. This means that long memory models do not actually have significant predictive power as it was postulated in the literature, since even the simplest regime-switching model outperforms them. Next, the quality of long memory estimation is not likely to lead to strong conclusions, such as market inefficiency. Hence, we are more inclined to think that it is actually regime-switching nature of the process, not a long-range dependence that drives the market dynamics.
Concluding remarks
In this paper we empirically investigate validity of the long memory specification for BRIC countries. It has been noted in the literature that processes with structural breaks are often confused with those exhibiting long memory. We suggest assessing the validity of the model by its predictive power. We estimate ARFIMA and Markov Switching models, and build series of one-step ahead forecasts. Predictive powers of the models are compared using Granger Newbold and Diebold Mariano tests. We demonstrate that even the simplest specification of MS-model shows superior predictive power to that of ARFIMA for all the countries. This evidence raises doubts concerning the use of the long memory presence as an indicator of good time series prediction and a sign of market inefficiency, while the regime-switching nature of the process is more likely to be present in the data. However, one should carefully interpret the conclusions of this paper, since fractional integration is only one of the means to capture long-range dependence. Alas, most of other techniques do not allow for predictability of the series, and ARFIMA-model stays one the most popular means of analysis.
Despite poor predictive power, long range dependence has a clear advantage of representing the movements in a heterogeneous agents market, which is quite a realistic assumption. On the other hand, it is not clear what causes sudden changes in the regime of the model. We think that the solution might lie in the combination of agents heterogeneity and changes in that structure. For instance, a substantial negative shock could undermine the trust of long-term investors, and they temporarily abstain from transactions, so that the market is driven mostly by short-term investors. It seems that dynamic structure of market participants could perfectly reconcile heterogeneous agents framework and regime-switching behavior of the index. We believe that further research will be able to capture these effects.
References
An S., Bloomfield P. Cox and Reids Modification in Regression Models with Correlated Errors: Discussion paper. Department of Statistics. North Carolina State University. Raleigh, 1993.
Andrews D. Tests for Parameter Instability and Structural Change with Unknown Change Point // Econometrica. 1993. 61. 4. . 821 856.
Andrews D., Ploberger W. Optimal Tests When a Nuisance Parameter Is Present Only under the Alternative // Econometrica. 1994. Vol. 62(6). . 1383 1414.
Bai J., Perron P. Estimating and Testing Linear Models with Multiple Structural Changes // Econometrica. 1998. 66. . 47 78.
Brock W.A., Hommes C.H. Heterogeneous Beliefs and Routes to Chaos in a Simple Asset Pricing Model // Journal of Economic Dynamics and Control. 1998. Vol. 22(8-9). . 1235 1274.
Brown R.L., Durbin J., Evans J.M. Techniques for Testing for Constancy of Regression Relationships Over Time // Journal of the Royal Statistical Society B. 1975. 37. . 149 163.
Cajueiro D.O., Tabak B.M. Testing for Predictability in Equity Returns for European Transition Markets // Economic Systems. 2006. Vol. 30(1). . 56 78.
Choi K., Zivot E. Long Memory and Structural Changes in the Forward Discount: An Empirical Investigation // Journal of International Money and Finance. 2007. 26. . 342 363.
Dacorogna M., Gencay R., Muller U. et al. An Introduction to High Frequency Finance. San Diego, CA: Academic Press, 2001.
Dacorogna M., Aste T., Di Matteo T. Long-term Memories of Developed and Emerging Markets: Using the Scaling Analysis to Characterize their Stage of Development // Journal of Banking & Finance. 2005. Vol. 29. . 827 851.
Diebold F.X., Inoue A. Long Memory and Regime Switching // Journal of Econometrics. 2001. 105. . 131 159.
Diebold F., Mariano HIdeǯǓo^G^8h;OJQJ\mH sH ,jh5;CJOJQJUmHnHu h5;CJOJQJmH sH h5;CJOJQJmH sH $h5;CJ OJQJ^JmH sH 6jh0JCJ EHH*OJQJU^JaJmH sH /h0J)B*CJ NHOJQJ^JmH phsH +h0J)B*CJ OJQJ^JmH phsH h5CJOJQJmH sH hCJmH sH hCJmH sH
HIIEkdT$$Ifl0]4
lalEkd$$Ifl0]4
lal<d$If] dL$If
dL$If]$If!
$^`a$
7$8$H$`
d7$8$H$`d<7$8$H$`d`
d<```$`a$? @ 9
:
:
efUV˻uauShB*CJmH phsH 'h@B*CJNHaJmH phsH #h@B*CJaJmH phsH #h@B*CJNHmH phsH h@B*CJmH phsH #h@B*CJNHmH phsH h@B*CJmH phsH h@B*CJph'hB*CJOJQJ^JmH phsH (h5@CJOJQJ\^JmH sH GH89'(sttu$%opsghST9:<E
˿˭˭˭˭˝˭˭˭˭˭˝˭˭˭˭˭˝˭˝˝˭#h@B*CJNHmH phsH h@B*CJmH phsH #h@B*CJNHmH phsH h@B*CJphh@B*CJmH phsH 'hB*CJOJQJ^JmH phsH h@B*CJmH phsH 6OF"J"h"j"""2#b#$s$$%1%<&'0'$`a$
d<`
$`a$$^`a$``d`
d<`%h6@B*CJ]mH phsH $jh@B*CJEHUphjַN
hB*UVph jh@B*CJUph$jh@B*CJEHUph(jh@B*CJUmH phsH h@B*CJph"h@B*CJ]mH phsH "h6@B*CJmH phsH #h@B*CJNHmH phsH h@B*CJmH phsH ++++++++++D,E,G,,,,,,,,,˿˿~n[F(jh@B*CJUmH phsH $jh@B*CJEHUphjַN
hB*UVphhB*CJNHmH phsH hB*CJmH phsH $jh@B*CJEHUphjxַN
hB*UVphh@B*CJph jh@B*CJUph%h6@B*CJ]mH phsH h@B*CJmH phsH ,,,,, -A-B-T--------------ݶݤݔs`K8K$jh@B*CJEHUph(jh@B*CJUmH phsH $j5h@B*CJEHUphjXN
hB*UVph jh@B*CJUphh@B*CJmH phsH #h@B*CJNHmH phsH "h6@B*CJmH phsH (jh@B*CJUmH phsH h@B*CJmH phsH $jfh@B*CJEHUph,T--..<.=..u///000x11S2o2d`$`a$$`a$
d<`
^``d`<`d`--..<.=...2/3/F/G/H/I/J/Q/R/S/U/u/v//ξޠޏlWDWޏ8h@B*CJph$j\^h@B*CJEHUph(jh@B*CJUmH phsH $jl[h@B*CJEHUphjN
hB*UVph jh@B*CJUph#h@B*CJNHmH phsH h@B*CJphhB*CJ^JmH phsH h@B*CJmH phsH h@B*CJmH phsH "h6@B*CJmH phsH //////////////////////̞{̞fSf̞Cj`طN
hB*UVph$jh@B*CJEHUph(jh@B*CJUmH phsH $jh@B*CJEHUphjYطN
hB*UVphh@B*CJmH phsH "h5@B*CJmH phsH h@B*CJph jh@B*CJUph$jh@B*CJEHUphjطN
hB*UVph////000
000000000000000000001ǷǷtdQtܷ$j!h@B*CJEHUphjطN
hB*UVphh@B*CJph"h@B*CJ]mH phsH "h6@B*CJmH phsH $jh@B*CJEHUphh@B*CJmH phsH (jh@B*CJUmH phsH jh@B*CJUph$jh@B*CJEHUph111 1
111111121314151:1;1<1=1>1@1A1E1_1̷pppp]Mh@B*CJmH phsH %h6@B*CJ]mH phsH "h@B*CJ]mH phsH "h6@B*CJmH phsH $jo(h@B*CJEHUphh@B*CJmH phsH (jh@B*CJUmH phsH jh@B*CJUph$j%h@B*CJEHUphjطN
hB*UVph_1`1a1x112232o2p222222222222ͻͩ͘|i͘YF$jah@B*CJEHUphjٷN
hB*UVph$j]h@B*CJEHUphjfٷN
hB*UVphh@B*CJph jh@B*CJUph#h@B*CJNHmH phsH "h5@B*CJmH phsH h@B*CJmH phsH h@B*CJmH phsH #h@B*CJNHmH phsH o223233333355U77888::
d<`d`
&Fd<^`gd
d<```$`a$d`$`a$2222222233.3/3031323S3T33333۵ۤu۵eQEh@B*CJph'hB*CJOJQJ^JmH phsH h@B*CJmH phsH $jh@B*CJEHUphjٷN
hB*UVphh@B*CJph jh@B*CJUph%h6@B*CJ]mH phsH $jCeh@B*CJEHUphh@B*CJmH phsH (jh@B*CJUmH phsH 3333.4/466661828888q9r999:::::::::ͻͻͻͻͧͻͻ͏qq_qRh5@CJmH sH #h5@CJ\mHnHsH ujh5@CJU\h5@CJ\mH sH h@B*aJphh@B*CJph'hB*CJOJQJ^JmH phsH #h@B*CJNHmH phsH h@B*CJmH phsH #h@B*CJNHmH phsH h@B*CJmH phsH :::::::;tc$<<$If`a$<<$If`
&`EkdG$$Ifl0]J
4
lal$"$If^"`a$$If^`
`:::::::::;;7;8;;;;;<<6<7<S<T<V<p<q<<<<==!="=========== >>&>(>J>K>L>N>W>X>˹˹ˬˬˬˬˬˬˬˬˬːˬˬˬˬˬˬˬ˄˄h@B*CJphh@B*CJmH phsH h@CJmH sH h@CJH*mH sH #h@B*CJNHmH phsH h@B*CJmH phsH hCJmH sH h6@CJ]mH sH h@CJmH sH 4;;; ;=;V;wfff$<<$If`a$<<$If`ykd$$Ifl40b`_
t0]44
laqf4V;W;X;~;;;_QQQQ..$If`kdO$$Ifl4\*
Qb ''
t0]44
laqf4;;;;<0<bTCCC$..$If`a$..$If`kd$$Ifl\*
Qb''
t0]44
laq0<1<2<O<l<<bTCCC$..$If`a$..$If`kd$$Ifl\*
Qb''
t0]44
laq<<<<<<bTCCC$..$If`a$..$If`kdӮ$$Ifl\*
Qb''
t0]44
laq<<<==8=bTCCC$..$If`a$..$If`kd$$Ifl\*
Qb''
t0]44
laq8=9=:=W=t==bTCCC$..$If`a$..$If`kd$$Ifl\*
Qb''
t0]44
laq======bP???$..$If`a$..$If]`kdv$$Ifl\*
Qb''
t0]44
laq===>!><>bTCCC$..$If`a$..$If`kdW$$Ifl\*
Qb''
t0]44
laq<>=>D>E>F>a>bTCCC$..$If`a$..$If`kd8$$Ifl\*
Qb''
t0]44
laqa>b>l>>>>bP???$..$If`a$..$If]`kd$$Ifl\*
Qb''
t0]44
laqX>b>l>??>???_?a???????????@@2@3@@@@@@@@@@@AAA%A'A)A>AHAvA|AAAbBdBBBBBBBBBCC-C2CKCMCoCpCCCCCDD;DhCJmH sH h@B*CJphh@CJmH sH h@CJH*mH sH h@B*CJmH phsH h@B*CJmH phsH E>>>>>?bTTTT..$If`kd$$Ifl\*
Qb''
t0]44
laq???=?^?~?bTCCC$..$If`a$..$If`kd۵$$Ifl\*
Qb''
t0]44
laq~??????bTCCC$..$If`a$..$If`kd$$Ifl\*
Qb''
t0]44
laq????@1@bTCCC$..$If`a$..$If`kdI$$Ifl\*
Qb''
t0]44
laq1@2@3@O@k@@bP???$..$If`a$..$If]`kd$$Ifl\*
Qb''
t0]44
laq@@@@@@bTCCC$..$If`a$..$If`kd$$Ifl\*
Qb''
t0]44
laq@@@A!A=AbTCCC$..$If`a$..$If`kdn$$Ifl\*
Qb''
t0]44
laq=A>AHA^ArAAbP???$..$If`a$..$If]`kd%$$Ifl\*
Qb''
t0]44
laqAAAAAAbTTTT..$If`kdܺ$$Ifl\*
Qb''
t0]44
laqAAAB**
`d`
d<``kd'$$Ifl\*
Qb''
t0]44
laqHHHHHIIRISIjIkIlImInIoIIIIIIIII˿scL<jrѸN
hB*UVph,jh@B*CJEHUmH phsH jҸN
hB*UVph#j
h@CJEHUmH sH jgѸN
hUV(jh@B*CJUmH phsH h@B*CJmH phsH hCJmH sH h@CJmH sH h5@CJmH sH h5@CJ\mH sH h@B*aJphh@B*CJphHHIII$I8IJI{{{$22$If`a$
&`Ekd$$Ifl0]J
4
lal$"$If^"`a$$If^`JIKIRInII|ndd
$If`9$If^9`kd@$$IfFS
x
t0X4aIIIII|ndd
$If`9$If^9`kdK$$IfFS
x
t0X4aIIIIIIIIIIIIIIIIIJJJJJJ(J)JĴčvfO?jUӸN
hB*UVph,jh@B*CJEHUmH phsH jҸN
hB*UVph,jh@B*CJEHUmH phsH j5ҸN
hB*UVph,jh@B*CJEHUmH phsH jҸN
hB*UVphh@B*CJmH phsH (jh@B*CJUmH phsH ,jh@B*CJEHUmH phsH IIII J|ndd
$If`9$If^9`kdO$$IfFS
x
t0X4a J
JJ,JHJ|ndd
$If`9$If^9`kd\$$IfFS
x
t0X4a)J*J+J,J-JDJEJFJGJIJJ?K@K M
MMMMĴč}kQE7h@B*CJaJphh@B*CJph3jh0J@B*CJEHH*UmH phsH #h@B*CJNHmH phsH h@B*CJmH phsH h@B*CJmH phsH ,jh@B*CJEHUmH phsH jӸN
hB*UVphh@B*CJmH phsH (jh@B*CJUmH phsH ,jh@B*CJEHUmH phsH HJIJFKuMMMM|qqh^N$If^`
`d`
d<`kdU$$IfFS
x
t0X4aMMMMM N
NNNNNNNNN$N%NLNMNNNzP{P|PPPPPPξtfXh@CJNHmH sH h@B*CJaJphh@B*CJph#h@B*CJNHmH phsH h@B*CJmH phsH h@B*CJmH phsH h@B*CJphh@B*CJmH phsH hCJmH sH h@CJmH sH h5@CJmH sH h5@CJ\mH sH MMMMMMMMM$<<$If`a$
&`Ekd$$Ifl0]J
4
lal$"$If^"`a$MM NNNOA,,$<<$If]`a$<<$If`kds$$Iflr[
b
t0]44
laqN#N*N+N6N:,<<$If`kd*$$Iflr[
b
t0]44
laq$<<$If]`a$6N=N@NCNKNLN:kd$$Iflr[
b
t0]44
laq$<<$If]`a$LNMN{P|PPPPPPPPP}sbbbb$<<$If`a$
&`Ekd$$Ifl0]J
4
lal$"$If^"`a$$If^`
`d``PPPPPPPQQGQHQuQvQQQQQQRRR R!RKRLRtRuRRRRSiSjSBUCUǷǷǷǷǷǷǷǷǷǷǷǷueh@B*CJmH phsH #h@B*CJNHmH phsH h@B*CJmH phsH h@B*CJmH phsH
hCJhCJmH sH h@B*CJmH phsH %h6@B*CJ]mH phsH hCJmH sH h@CJmH sH h@CJmHnHsH u"PPPP>0FF$If`kd$$Iflr,y
b'MNMN
t0]44
laq$<<$If`a$PPQQQQ:kd$$Iflr,y
b'MNMN
t0]44
laq$FF$If]`a$Q Q)Q2Q=QFQ$FF$If]`a$FF$If`FQGQNQWQ`QOA,,$FF$If]`a$FF$If`kd$$Iflr,y
b'MNMN
t0]44
laq`QkQtQuQ{Q:,FF$If`kde$$Iflr,y
b'MNMN
t0]44
laq$FF$If]`a${QQQQQQ:kd8$$Iflr,y
b'MNMN
t0]44
laq$FF$If]`a$QQQQQQ$FF$If]`a$FF$If`QQQQQOA,,$FF$If]`a$FF$If`kd$$Iflr,y
b'MNMN
t0]44
laqQQQQR:/
$P`a$kd$$Iflr,y
b'MNMN
t0]44
laq$FF$If]`a$RRRRRR R>kd$$Iflr,y
b'MNMN
t0]44
laq$<<$If`a$ R&R.R7RARJR$<<$If]`a$<<$If`JRKRQRZRaROA,,$<<$If]`a$<<$If`kd$$Iflr,y
b'MNMN
t0]44
laqaRjRsRtRzR:,<<$If`kds$$Iflr,y
b'MNMN
t0]44
laq$<<$If]`a$zRRRRRR:kdF$$Iflr,y
b'MNMN
t0]44
laq$<<$If]`a$RRRRRR$<<$If]`a$<<$If`RRRBUCUVUOD;5/``d`
d<`kd$$Iflr,y
b'MNMN
t0]44
laqCUVUWUUUUUVVWW%Y&Y[[[1\3\4\\\\\\] ]!],]-]n]o]]]\^^^^^^^ϽϽϽϽϽϭϭϽϽϏsϽϽϽaϏ"h@B*CJ]mH phsH hB*CJ^JmH phsH h@B*CJphh@B*CJph#h@B*CJNHmH phsH h@B*CJmH phsH #h@B*CJNHmH phsH h@B*CJmH phsH h@B*CJph'hB*CJOJQJ^JmH phsH &VUWUY ]!],]-]]^`adbcdVfJgii012:4@5d<7$8$H$`
d<`
^`d`
d<``^d_f___````bbHbJbccddee>f@fffthvh~hhhhiij0h0j0001L111N2P22234"4$4(5*5@5555p6r66677.9:<h@B*CJmH phsH "h@B*CJ\mH phsH Uh@B*CJph#h@B*CJNHmH phsH h@B*CJmH phsH ?R.S. Comparing Predictive Accuracy // Journal of Business & Economic Statistics. 1995. Vol. 13(3). . 253 263.
Ding Z., Granger C.W.J. Varieties of Long Memory Models // Journal of Econometrics. 1996. Vol. 73(1). . 61 77.
Eom C., Oh J., Jung W.-S. Relationship Between Efficiency and Predictability in Stock Price Change // Physica A. 2008. 387. . 5511 5517.
Granger C., Hyung N. Occasional Structural Breaks and Long Memory with an Application to the S&P 500 Absolute Stock Returns // Journal of Empirical Finance. 2004. 11. . 399 421.
Granger C., Newbold P. Forecasting Transformed Series // Journal of the Royal Statistical Society. 1976. Series B. 38. . 189 203.
Hamilton J.D. Rational-Expectations Econometric Analysis of Changes in Regime: An Investigation of the Term Structure of Interest Rates // Journal of Economic Dynamics and Control. 1988. 12. . 385 423.
Jamdee S., Los C. Long Memory Options: LM Evidence and Simulations // Research in International Business and Finance. 2007. Vol. 21(2). . 260 280.
Kramer W., Sibbertsen P. Testing for Structural Change in the Presence of Long Memory: Technical Report 31/2000. SFB 471. University of Dortmund, 2000.
Kim C., Nelson J. State Space Models with Regime Switching: Classical and Gibbs Sampling Approaches with Application. Massachusetts: MIT Press, 1999.
LeBaron B. Agent Based Computational Finance: Suggested Readings and Early Research // Journal of Economic Dynamics and Control. 2000. 24. @5678.9<==? @r@t@x@z@~@@@@@@@@@@&`$hh1$]h`ha$&``$`
d<` . 679 702.
Mandelbrot B., Wallis J. Noah, Joseph and Operational Hydrology // Water Resources Research. 1968. Vol. 4. . 909 918.
Rogers L. Arbitrage with Fractional Brownian Motion // Mathematical Finance. 1997. 7. . 95 105.
Sowell F. Maximum Likelihood Estimation of Stationary Univariate Fractionally Integrated Time Series Models // Journal of Econometrics. 1992. Vol. 53. . 165 188.
We gratefully acknowledge financial support through State University Higher School of Economics scientific fund, grant !08-04-0043.
For ARFIMA-model we use EML-estimates.
PAGE 290
PAGE 279
PAGE 404
<<<<<<<<<==8=L=========Z>\>f>>>???ӵӡӡӏӡӵ{iW#h@B*CJaJmH phsH #h@B*CJNHmH phsH 'jh@B*UmHnHphu#h@B*CJNHmH phsH &h0J,6@B*CJmH phsH h@B*CJph"h@B*CJ]mH phsH h@B*CJmH phsH h@B*CJmH phsH h@B*CJph???@ @"@r@t@v@x@z@|@@@@@@@@@@@@@@@@@@@@@@@@@@@@˹ӯӯӒ}k#h@B*CJaJmH phsH h0JCJmHnHuh0JCJjh0JCJUh]0JmHnHu
h0Jjh0JU#h@B*CJaJnHphtHjhUhhmH sH hCJmH sH hH*mH sH jh0JEHH*U&@@@@@@@@@@
d<`$`a$
$h1$]ha$ <00,P&P . A!"#$ %CDpR$$Ifl!vh#v#v:Vl554alR$$Ifl!vh#v#v:Vl554al]Dd
<
TJ
CA?"2Im9`6H,`!Im9`6H` HbXJaxڥSoQyRF-bZ{#)Rl`QJk$Xz0?I^îL{S$[,
19H<>ǖ˺>XXёdDb3FG\tspLi߿fX8lB'ⴿ5W~!ZZdX_)+ugOH?ﰅLĞ;A
&hmm}>\\D+jZQ|iCo@7yDr-K+zs-eK0A1Ue9'EBf/ǲ@2TqRH%A!HJ9ȅzt^Ў1֎>!wƦGd+{Z;>VLj+r' I!?~)so%О!wGoyΚ
#ZnxwP>n$~HEDd
?
3A"V?P?
12?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~Root Entry FPћVZData
WordDocument>HObjectPool'VPћV_1320670290FVVOle
CompObjiObjInfo
"%&'()*+,-./256789:=@ABCDEFGJMNORUVWXYZ[\_bcdefilmnopsvwxy|
FMathType 5.0 EquationMathType EFEquation.DSMT49qgxPDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!'_APG_APAPAE% _AC_AE*_HA@AHA_D_E_E_A
$(L)(1"-Equation Native _1320670334FVVOle
CompObj
iL)d
(yt
"-m)==Q(L)et
,
FMathType 5.0 EquationMathType EFEquation.DSMT49qoxPDSMT5WinAllBasicCodePagesObjInfo
Equation Native _1320670368FVVOle
Times New RomanSymbolCourier NewMT Extra!'_APG_APAPAE% _AC_AE*_HA@AHA_D_E_E_A
(1"-L)d
x(t)==e(t),d"("-1/2,1/2),
FMathType 5.0 EquationMathTyCompObjiObjInfoEquation Native d_1320670416 FFVFVpe EFEquation.DSMT49qHxPDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!'_APG_APAPAE% _AC_AE*_HA@AHA_D_E_E_A
e(t)"-i.i.d.0,se
2
().l="00144189"/>**